Separation of Foreground and Background in Medical Images

1. A method, executed by one or more processors, the method comprising: computing a plurality of binarization scores for an image using a corresponding plurality of binarization threshold values; determining a selected binarization threshold value based on the plurality of binarization scores; processing the image using the selected binarization threshold value into a first binary image; determining a transition pixel count and a non-transition pixel count for the first binary image, wherein the transition pixel count comprises a count of pixels of differing binary values that are adjacent to each other in the first binary image and the non-transition pixel count comprises a count of pixels of equal binary values that are adjacent to each other in the first binary image; determining a first binarization score, wherein: the first binarization score is based on the non-transition pixel count and the transition pixel count for the first binary image, the first binarization score corresponds to a ratio between a non-transition pixel count function and a transition pixel count function for the first binary image, and the transition pixel count function corresponds to a sum of a black-to-white pixel transition count and a white-to-black pixel transition count for the binary image; and processing the first binary image using the first binarization score into a second binary image.

2. The method of claim 1 , wherein the non-transition pixel count function corresponds to a product of an adjacent black pixel count and an adjacent white pixel count for the binary image.

3. The method of claim 1 , wherein the transition pixel count comprises a count of black-to-white pixel transitions and/or a count of white-to-black pixel transitions.

4. The method of claim 1 , wherein the transition pixel count corresponds to a selected dimension of the image.

5. The method of claim 1 , wherein each binarization score of the plurality of binarization scores is computed by effectively partitioning a gray level co-occurrence matrix into quadrants.

6. The method of claim 5 , wherein the quadrants correspond to adjacent black pixels, adjacent white pixels, black-to-white pixel transitions, and white-to-black pixel transitions.

7. The method of claim 1 , wherein each binarization score of the plurality of binarization scores is computed by summing rows and columns of a gray level co-occurrence matrix.

8. The method of claim 1 , wherein each binarization score of the plurality of binarization scores is inversely related to the transition pixel count.

9. The method of claim 1 , wherein the selected binarization threshold value corresponds to the one or more of a local maxima, a global maxima, and a maximum positive slope for the plurality of binarization scores.

10. The method of claim 1 , wherein each binarization score of the plurality of binarization scores is proportional to the transition pixel count.

11. The method of claim 1 , where in the selected binarization threshold value corresponds to one or more of a local minima, a global minima, and a maximum negative slope for the plurality of binarization scores.

12. The method of claim 1 , wherein the image is a mammogram.

13. A computer program product comprising: one or more computer-readable storage media that are not transitory signals per se and program instructions stored on at least one of the one or more computer-readable storage media, the program instructions comprising instructions for: computing a plurality of binarization scores for an image using a corresponding plurality of binarization threshold values; determining a selected binarization threshold value based on the plurality of binarization scores; processing the image using the selected binarization threshold value into a first binary image; determining a transition pixel count and a non-transition pixel count for the first binary image, wherein the transition pixel count comprises a count of pixels of differing binary values that are adjacent to each other in the first binary image and the non-transition pixel count comprises a count of pixels of equal binary values that are adjacent to each other in the first binary image; determining a first binarization score, wherein: the first binarization score is based on the non-transition pixel count and the transition pixel count for the first binary image, the first binarization score corresponds to a ratio between a non-transition pixel count function and a transition pixel count function for the first binary image, and the transition pixel count function corresponds to a sum of a black-to-white pixel transition count and a white-to-black pixel transition count for the binary image; and processing the first binary image using the first binarization score into a second binary image.

14. The computer program product of claim 13 , wherein the non-transition pixel count function corresponds to a product of an adjacent black pixel count and an adjacent white pixel count for the binary image.

15. A computer system comprising: one or more computer processors; one or more computer-readable storage media; and program instructions stored on at least one of the one or more computer-readable storage media for execution by at least one of the computer processors, the program instructions comprising instructions for: computing a plurality of binarization scores for an image using a corresponding plurality of binarization threshold values; determining a selected binarization threshold value based on the plurality of binarization scores; processing the image using the selected binarization threshold value into a first binary image; determining a transition pixel count and a non-transition pixel count for the first binary image, wherein the transition pixel count comprises a count of pixels of differing binary values that are adjacent to each other in the first binary image and the non-transition pixel count comprises a count of pixels of equal binary values that are adjacent to each other in the first binary image; determining a first binarization score, wherein: the first binarization score is based on the non-transition pixel count and the transition pixel count for the first binary image, the first binarization score corresponds to a ratio between a non-transition pixel count function and a transition pixel count function for the first binary image, and the transition pixel count function corresponds to a sum of a black-to-white pixel transition count and a white-to-black pixel transition count for the binary image; and processing the first binary image using the first binarization score into a second binary image.

16. The computer system of claim 15 , wherein the non-transition pixel count function corresponds to a product of an adjacent black pixel count and an adjacent white pixel count for the binary image.